Recent years have seen large advances in new artificial intelligence (AI) techniques for improving the current status of medical imaging, and specifically, magnetic resonance imaging (MRI).

In this work, we accelerate and improve the reconstruction of multi-shot diffusion-weighted MRI by an unrolled pipeline, in which the presumed regularization term is replaced by a U-Net.

Despite advances in accelerating MRI scans, diagnostic knee MRI protocols typically require upwards of 30 minutes of scanner time, which fundamentally limits patient throughput.

Acquisition of diffusion-weighted images along multiple directions is needed for deriving microstructural metrics from diffusion models, and may also provide valuable information for some clinical applications.

Deep learning based super-resolution (SR) is a computer vision method that can enhance the resolution of low-resolution images, which has recently been applied to MRI.

We apply automatic differentiation (widely used in deep learning) and a simple Bloch simulation to efficiently optimize any sequence that can be simulated.

Clinical knee MRI examinations typically utilize 2D imaging methods with thick slices, requiring upwards of 20-25 minutes of scan time, and yet not producing any quantitative information.

Total joint replacement is a common surgery to treat end-stage joint pain, which is projected to exceed 4.

Diffusion-weighted MRI is a widely used technique in neuroscience research, as well as for many clinical applications.